Jenny Love, UCL Energy Institute
In my previous post here, I described the process of PhD data collection and some of the odd things which can happen when carrying it out. After publishing that post I went quiet for a bit to concentrate on the small matter of turning the data into a 90,000 word thesis. Having done that, and then recovering by alternating between sleep and eating nutella out of the jar, I am ready to face the world again. So in this post and the next one I’ll be describing some of my findings.
In this post I’ll talk about what light was shed on the complexities of occupant behaviour in the context of retrofit: how occupants change their heating behaviour afterwards, and why internal temperatures increase. In the next post I’ll report some implications of this concerning whether our current retrofit strategy for social housing goes far enough.
What was I trying to find out?
There is a general concern that retrofit, here referring to insulation of dwellings, doesn’t save as much energy as it should for all the effort of wrapping a house up in a blanket. That’s because it is suspected that instead of what is ‘supposed’ to happen (i.e. the occupants keep the heating at the same level after the retrofit and save energy), occupants will take this chance to heat to higher temperatures or for longer, and not save much energy after all. This is especially expected to happen in ‘fuel poor’ households: those who before retrofit struggle to heat their home to the temperature they would want, for whom insulation might mean they can be warmer.
As I described in more detail in the last post, instead of looking at just outcomes of retrofit (i.e. these people saved X kWh of energy, or it was X degrees warmer in their house afterwards), I looked also at what actually happened in the homes: what did occupants think the retrofit was all about; how had they changed their use of heating since; were they using their home differently and was this requiring more or less energy? So the sort of things I was measuring were: their use of radiators, air temperature and humidity, and use of space; I also interviewed the occupants before and after.
Variety in occupant reactions
When you wrap up a building in a big pink jumper (see photo below), it will lose heat less quickly afterwards, and so its average temperature will increase without the occupants doing anything.
But how occupants then react to this natural temperature increase is different in different households. I found three types of occupant reaction, ranging from occupants practically eliminating their use of heating to occupants using more hours of heating:
1. Temperature increase from the building, counteracted by occupants
Two of the case study households turned down the heating so much that the temperature went down after retrofit. However, they reported feeling warmer. How could both of these phenomena have come about? Here’s what I could gather from my data…
Both of these households had an income cut around the time of the retrofit and were really struggling for money. Also, before the retrofit they were both expecting the retrofit to lead to a warmer house with less heating needed. This could explain why they turned the heating down so much. But why did they feel warmer? I looked in the air temperature data: had the particular rooms they used got warmer at the times they used them? Or had the daily minimum temperature they experienced increased, say, when they got up in the morning? Neither of these had occurred. I had to conclude that there must be other comfort variables at work, like radiant temperature.
2. Temperature increase from the building, no change in occupant behaviour
These occupants carried on using the heating in pretty much the same way as before: they didn’t turn anything down, and they didn’t turn anything up. As was mentioned above and will be further explained below, this still leads to an increase in internal temperature.
3. Temperature Increase from the building, then occupants used more heating
Two different and interesting processes were seen in households who used the heating more after retrofit.
In one flat, the occupant didn’t really bother with the central heating before the retrofit since the building was so leaky that it didn’t seem to make a difference. He heated the living room with a gas fire and stayed in there. After the retrofit he started using the central heating since it now actually did something.
In another case, even though the insulation made the house warmer, different rooms warmed up by different amounts. This had the effect of making the occupants’ bedroom feel cold, as it hadn’t warmed up as much as some other rooms. So they started heating the whole house all evening, so that their bedroom would be warm enough by the time they wanted to use it.
Turning heating up, down, the same…does that mean that anything can happen?
Yes. I will argue in my next post that with this type of ‘shallow’ retrofit (10 cm of insulation), it’s very difficult to predict the outcomes since you leave a lot of room for different outcomes to be possible, especially when new occupants move in. However we only started to see a glimpse of the range of possible outcomes. With the same people living there after retrofit as before, occupants probably aren’t going to massively increase their energy use after retrofit so we probably aren’t going to see those kind of outcomes.
Also, just because occupants reacted in a variety of ways, doesn’t mean that how they reacted totally determined the outcome. As we’ve seen, the building theoretically has a lot to do with it. We can try to quantify its influence in a few ways…
In most houses there was a temperature increase. What was it mostly caused by?
The answer is not the occupants but the building itself. How can I know this? By looking at when it occurred…
Firstly, most of the temperature increase compared to the previous year occurred when the heating was off. That is, the times when the heating was off, post-retrofit, were warmer than the times the heating was off pre-retrofit. I had a fancy equation to calculate how much of the temperature increase occurred during unheated hours, and it turned out to be 77-87% across the houses.
Within this, quite a bit of the temperature increase happened at night. It’s possible to see the houses cooling down slower at night, when the heating was off, whilst the occupants are fast asleep and therefore not thinking, ‘I know, I’ll increase the temperature in my dwelling’.
In some houses, the hours in which the heating was on got warmer after retrofit (accounting for 5%-23% of the temperature increase). This wasn’t due to people turning up the thermostats. It was either that the thermostat was at a sensible setting and the building was too leaky to get that warm before retrofit, or the thermostat was at a non-sensible setting like 30C (for whatever reason) and the heating system tried its hardest but still couldn’t reach it.
What to make of all this
The first point is that if the temperature increases in someone’s home after retrofit, it’s not necessarily their fault or their intention. To get the temperature not to increase, occupants would have to shorten their daily heating period by quite a few hours. Even when they increase their hours of heating, most of the temperature increase is still attributable to the building cooling down more slowly. So we can stop occupant-blaming.
Secondly, there was quite a large range in terms of how occupants reacted, and whether they turned heating up or down, which makes it difficult to predict outcomes. However, this was a small range compared to what could have happened. I haven’t explained this statement yet, as I explore this further in the next post, where I talk about the effect of new tenants moving in over time – and how the current way we do retrofit means we can’t guarantee energy savings afterwards.
That’s all for now – but feel free to get in touch with me if you would like any more detail on the sort of mechanisms I uncovered or if you have any questions: